INDEX

Explanations

statements suggesting conclusions or observations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating perception or opinion about situations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orem

-0.82

venge

-0.81

otos

-0.81

ests

-0.80

rouse

-0.80

andise

-0.79

atching

-0.75

ategory

-0.72

iling

-0.72

umbn

-0.72

POSITIVE LOGITS

 doubtful

0.81

 unlikely

0.80

 unclear

0.77

 unfair

0.76

 impossible

0.75

 inevitable

0.73

 abundantly

0.71

 probable

0.70

 obvious

0.70

 Tycoon

0.69

Activations Density 0.054%

statements suggesting conclusions or observations

phrases indicating perception or opinion about situations

No Comments

No Known Activations